Bundles and upsells for digital products
Doubling your traffic is months of work. Doubling what each buyer spends is an afternoon, and almost nobody does it.
The short version
- A buyer is worth several times a visitor. They have already decided to trust you.
- The order bump is the highest-return thing you can add. One checkbox, 20–40% take rate.
- Bundle below the sum, above the dearest item. That gap is the whole mechanism.
- Upsell the next problem, not more of the same thing.
- Two products beat one, and four beat two. Stop well before ten.
Why is the second sale so much easier?
Because the hard part is already done. Getting a stranger to trust you enough to enter a card is most of the work in selling anything; someone who just did that is, for a few minutes, the warmest audience you will ever have. Selling them something else costs you nothing in reach.
The arithmetic makes the point better than the argument. Take a hundred visitors, a 3% conversion rate and a $29 product: that is three sales and $87. Now add an order bump at $19 that 30% of buyers take, and an upsell at $49 that 20% take:
| Without | With | |
|---|---|---|
| Visitors | 100 | 100 |
| Buyers | 3 | 3 |
| Core product | $87 | $87 |
| Order bump ($19, 30%) | None | $17 |
| Upsell ($49, 20%) | None | $29 |
| Total | $87 | $133 |
Same traffic, same product, same conversion rate: 53% more revenue. And unlike a traffic increase, it does not need to be earned again next month. This is why every experienced seller has a second thing to offer at the checkout and most first-time sellers have nothing.
What is the difference between a bump, an upsell and a bundle?
An order bump is a checkbox on the checkout page adding something small. An upsell is an offer shown immediately after payment. A bundle is several products sold together at one price, before either of those happens. They stack, and they do different jobs.
| When | Price | Take rate | |
|---|---|---|---|
| Order bump | On the checkout page | 20–40% of the main price | 20–40% |
| Upsell | Right after payment | 1–2× the main price | 5–20% |
| Downsell | After an upsell is declined | Half the upsell | 5–10% |
| Bundle | Instead of a single product | Below the sum of parts | Varies |
| Cross-sell | In the delivery email, days later | Anything | 2–5% |
Start with the order bump. It is one checkbox, it takes twenty minutes to set up on Gumroad or Payhip, it has the highest take rate of anything on the list, and it cannot cost you the main sale: the buyer has already committed by the time they see it.
What makes a good order bump?
Something small, cheap, and immediately obvious as a companion to what they are already buying. The test is whether it completes the job: if the main product is a template, the bump is the guide for filling it in. If it needs explaining, it is an upsell, not a bump.
- Price it at 20–40% of the main product. $9 against a $29 purchase feels like rounding; $29 against $29 is a second decision.
- One sentence and a checkbox. Nobody reads a paragraph at the checkout.
- Make it genuinely complementary. Video walkthrough, swipe file, extra templates, a checklist version.
- Never make it something the main product needed. If the core purchase feels incomplete without the bump, you have built resentment into your checkout.
That last rule is the one that separates this from the tactics people rightly dislike. A bump that adds is fine. A bump that unlocks what should have been included is a fee, and buyers correctly read it as one.
How do you price a bundle?
Below the sum of the parts and above the most expensive single item. That window is what makes the bundle the obvious choice, and the discount should be visible, stated as a number, next to the individual prices.
So three products at $29, $19 and $19 ($67 individually) bundle at somewhere between $39 and $49. Show the arithmetic on the page: “$67 separately, $47 together.” The buyer is then comparing two numbers instead of judging one, which is the same anchoring principle that makes a sales page work at all.
- Bundle things that are actually used together. A random assortment reads as clearing inventory.
- Keep selling the parts separately. The individual prices are what make the bundle price legible.
- Lead with the bundle on the page, with singles below it as the alternative.
- Do not discount past the point of insult. $67 of products for $19 tells the buyer what you think they were worth.
The bundle trap: if the bundle is much cheaper than the parts, nobody buys the parts, and you have simply cut your prices. The gap should reward buying everything, not punish buying one thing.
What should the upsell be?
The next problem, not more of the same. Someone who just bought a system for doing X does not want a second system for doing X. They want the thing that makes X faster, or the thing they will need once X is working. Depth over volume.
| They bought | Bad upsell | Good upsell |
|---|---|---|
| A template pack | More templates | The course on using them well |
| A guide to a platform | A guide to another platform | The done-for-you assets for that platform |
| A starter system | A second starter system | The advanced version, or ongoing support |
| A planner | A different planner | The full suite plus a walkthrough |
And keep it to one. A stack of three consecutive offers between payment and download is where a pleasant purchase turns into an obstacle course. One upsell, one optional downsell if declined, then the file: that is the whole sequence, and anything longer costs you goodwill worth more than the marginal sale.
How many products should you have?
Two is dramatically better than one, four is better than two, and beyond about six you are usually splitting attention rather than adding revenue. The right shape is a small ladder (an entry product, a main product, and something deeper), not a catalogue.
A ladder that works for most faceless sellers:
- $9–$19 entry. Its job is to turn a follower into a buyer. It rarely earns much and it makes everything after it easier.
- $29–$49 main product. The thing you actually sell, and the one the content points at.
- $79–$149 bundle or deep version. Where a meaningful share of the revenue comes from, bought by a small share of buyers.
The trap at the other end is building product four before product one has sold consistently. A second product does nothing for a first product nobody is reaching: that is still a traffic problem, and a launch sequence or a content engine fixes it, not more inventory.
One last thing worth doing on the day you set any of this up: put a link to your other products in the delivery email and on the thank-you page. It takes five minutes, it runs forever, and it is the cheapest cross-sell available: the buyer is opening that email precisely because they are excited about what they just bought.
Frequently asked questions
What is an order bump?
A checkbox on the checkout page offering a small, related add-on, usually 20–40% of the main price. Take rates of 20–40% are normal, and it is the highest-return thing you can add to a store.
Should the bundle be cheaper than buying separately?
Yes, but not dramatically. Price it below the sum of the parts and above your most expensive single item. Too steep a discount and nobody buys the individual products at all.
How many upsells is too many?
More than one, plus an optional downsell. Stacking three offers between payment and delivery converts slightly better on paper and costs you goodwill and refunds that are worth more.
Can I do this on Gumroad?
Yes. Gumroad, Payhip and most of the direct-sale platforms support bundles and some form of checkout add-on natively. You do not need a dedicated funnel tool for any of this.
What if I only have one product?
Split it or extend it. A workbook version, a video walkthrough, or a template set drawn from the same material gives you a bump without building anything genuinely new.
The system behind this, written down
Everything above is the map. The Income Loop is the work inside it: modules 0–6 from the problem you solve to the offer that pays for it, plus ten traffic paths: the deeper post banks, the content sales systems and the full software build sequence, in one place.
Get The Income LoopNot ready to pay for anything? The Basic Income Loop is free and includes a complete seven-day starter for Threads, Instagram or software, enough to find out which one suits you before spending anything.